Electronic trading refers to a trading system that matches orders by an electronic trading platform. Current examples of electronic trading platforms include the Chicago Mercantile Exchange (“CME”) Globex® trading system, the Chicago Board of Trade (“CBOT”) e-cbot, and Eurex, just to name a few. A trading system that matches orders electronically is also referred to herein as an “electronic exchange.” Exchanges like the CME and CBOT currently offer trading via open outcry in addition to electronic trading.
Traders link to an electronic exchange through a computer network. Then, using a front-end trading system, traders can participate in the electronic markets offered by these electronic exchanges. Electronic exchanges or independent companies like Trading Technologies® develop, deploy, and support front-end electronic trading systems that allow traders to access various electronic exchanges. Some of Trading Technologies® products include X_TRADER®, X_STUDY™, X_RISK®, and X_TRADER® API.
Presently, certain kinds of market data is multicast to the traders' computers over the network. As known in the art of computing, multicast refers to the delivery of information to many receivers at the same time in a single stream. Typically, the kinds of market data where multicast becomes useful is price information, such as the distribution of a product's current highest bid price, the current lowest ask price, the last trade price, and other types of price information. Currently, a trader may choose to subscribe from a list of hundreds, if not thousands, of different products, in which each product has its own multicast price feed. Generally speaking, the amount of market data, from which a trader may subscribe, will only increase in time as electronic exchanges add new products to their offerings and as the growth of trading volume for particular products grows.
According to one current network configuration type, a subscribing computer on a network must receive every, or nearly every, price feed that comes across the network, as a result of the price feeds having the same multicast address. This network configuration type requires that the subscribing computer filter the price feeds to retrieve the desired price data. The burden due to this undesirable filtering process is substantially increased as the number of computers subscribing to different price feeds are added to the same network and as the price feed for a given product increases as the trading volume for the product increases.
According to another current network configuration type, products are grouped onto several different channels, such that each channel has a certain bandwidth and a different multicast address. While this can reduce the number of price feeds a trader must subscribe to or filter, it works well if all the channels are equally busy and equally needed. Oftentimes, this is not realistic. Trading volume for some products may spike at different times, while the trading volume for other products might not spike at all. As a result, the network becomes under utilized at some times and over utilized at other times. Over utilization can lead to network failure, or at the very least, delayed price feeds. In a competitive environment like electronic trading, where every second (or even a fraction of a second) counts in seizing market opportunities, this rather inflexible network configuration is also an unacceptable long-term solution.
Therefore, what is needed is an improved network configuration that can be utilized going forward by systems that call for the distribution of data in environments such as electronic trading.